A major question is vision research is how human visual performance is limited by the architecture of the neural tissue which subserves it. Our goal is to establish a reliable and internally consistent anatomical database for comparing the anatomy of the human photoreceptor (PR) mosaic to visual function. In this proposal we continue to determine the extent to which age- related declines in visual function can be explained by loss of PR and ganglion cells (GC) in the retina. We use an unstained retinal whole mount and computer-assisted morphometry to characterize the spatial density of cones, rods, GC, and the size spectrum of GC somata in donor eyes over 70 years of age. We will also use these techniques to study a small number of eyes from elderly person with well documented visual function. Data on these cell types can serve as a base line for other studies of age-related maculopathy and glaucoma, major blinding disease of the elderly. We also propose to extend our work on the spatial distribution of cones in young retina to more detailed characterization of morphological substrates underlying sampling, sensitivity, and connectivity properties of the human cone mosaic. We will use Voronoi-based spatial statistics and Fast Fourier methods to analyze cone packing geometry as a function of eccentricity. We will investigate the spatial distribution of blue cones in the human retina using a recently developed antibody specific to the blue cone opsin. The ability of cones to gather light efficiently, is determined partly by inner and outer segment morphology. We plan to quantify this morphology using computer reconstruction and graphic display of optical sections through the long axis of clones. Finally, we will investigate the pattern of cone-cone connectivity in the peripheral retina by injecting a patch of adjacent cones in an in vitro preparation and examining the patterns of connections made by cone telodendria. We will also determine the projection of visual field onto the human GC layer by accounting for the lateral offset of foveal GC. Our previously collected data on the spatial density of GC can then be expressed in terms of cells/deg2 as a function of eccentricity in visual degrees. These data can be compared with the distribution of cones in the same eyes and with published estimates of the human cortical magnification factor.